Update app.py

app.py

@@ -3,6 +3,7 @@ import streamlit as st
 # working with sample data.
 import numpy as np
 import pandas as pd
+import faiss
 
 from sentence_transformers import SentenceTransformer
 
@@ -25,6 +26,43 @@ for sentence, embedding in zip(sentences, embeddings):
     st.write("")
 
 
+def get_embedding(text_content):
+    return model.encode(text_content)
+
+# Load the text file as knowledge
+knowledge_file = 'knowledge.txt'
+knowledge = []
+with open(knowledge_file, 'r', encoding='utf-8') as file:
+    for line in file:
+        knowledge.append(line.strip())
+
+# Create an index
+index = faiss.IndexFlatIP(300)  # Use Inner Product (IP) as similarity measure
+
+# Perform embedding for the knowledge texts and add to index
+embeddings = []
+for text in knowledge:
+    # Add your code here for text embedding (e.g., using word embeddings, sentence transformers, etc.)
+    embedding = get_embedding(text)
+    embeddings.append(embedding)
+embeddings = np.array(embeddings)
+index.add(embeddings)
+
+# Get user input for a question
+question = st.text_input("Enter your question: ")
+
+# Perform embedding for the question
+question_embedding = get_embedding(question)
+
+# Search index for the most similar content
+k = 5  # Number of results to retrieve
+D, I = index.search(np.array([question_embedding]), k)
+
+# Display the results
+st.write("Top {} similar content:".format(k))
+for i in range(k):
+    st.write("{}: {}".format(i+1, knowledge[I[0][i]]))
+
 st.title('My first app')
 
 st.write("Here's our first attempt at using data to create a table:")